537 - Urine Output Predicts Successful CRRT Discontinuation in Critically Ill Children

Publication Number: 4525.537

Julia W. McDonald, UPMC Childrens Hospital of Pittsburgh, Pittsburgh, PA, United States; Katherine L. Kurzinski, UPMC Childrens Hospital of Pittsburgh, Pittsburgh, PA, United States; Cassandra Formeck, UPMC Childrens Hospital of Pittsburgh, Pittsburgh, PA, United States; Jenna P. Harper, Children's Hospital of Pittsburgh of UPMC, Sewickley, PA, United States; Dana Fuhrman, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States; Jacqueline Ho, UPMC Childrens Hospital of Pittsburgh, Pittsburgh, PA, United States; Coleen J. Mcsteen, UPMC Childrens Hospital of Pittsburgh, Verona, PA, United States

Background: Continuous renal replacement therapy (CRRT) is a life-saving therapy commonly used to manage acute kidney injury and fluid overload in children. No clinical guidelines exist for CRRT discontinuation in recovering patients, leading to wide variability in pediatric practice regarding when to trial patients off of therapy.

Objective: The goal of this study is to describe predictors of successful CRRT discontinuation by examining the association of CRRT prescription and patient characteristics.

Design/Methods: We conducted a retrospective chart review of patients aged 3 months to 26 years who received CRRT at Children's Hospital of Pittsburgh from 2017-2022. Patients were excluded if they had a history of chronic kidney disease, congenital anomalies of the kidney and urinary tract, received ECMO during their CRRT course, received CRRT for toxic ingestion or hyperammonemia, or died during CRRT treatment. Successful CRRT liberation was defined as remaining off of CRRT or any dialysis modality for at least 7 days following the liberation attempt. Baseline comparisons between successful and unsuccessful liberation groups were performed using the Wilcoxon rank-sum test for continuous variables and Chi-square test for categorical variables. Analyses were conducted using Stata/SE 19.5. Variables with p < 0.2 in univariable analyses were entered into a multivariable logistic regression model to identify independent predictors of successful CRRT liberation.

Results: There were 108 patients who received CRRT and 61 patients were excluded from analyses. Of the 47 patients who met the inclusion criteria, 30 (63.8%) had successful liberation and 17 (36.2%) had unsuccessful liberation. There were no significant differences in age at ICU admission, sex, baseline serum creatinine, reason for ICU admission, illness severity, and CRRT prescription (Table 1). Patients with successful liberation had lower body surface area (BSA) and higher urine output (UOP) 24 hours prior to the liberation attempt. On multivariable analysis, UOP was the only variable that was independently associated with successful CRRT liberation (OR 591.4, 95% CI 9.2-37980.1, p< 0.001), after adjustment for BSA, baseline serum creatinine, blood flow rate, and anticoagulation type (Table 2).

Conclusion(s): Urine output in the 24 hours preceding the liberation attempt is strongly associated with successful CRRT discontinuation. Future directions include evaluating urinary biomarkers, such as fractional excretion of sodium and urea, and urine NGAL, and expanding analyses to include patients treated from 2022-2025 to validate and refine predictive models.

McDonald CV
McDonald CV.pdf

Table 2
Multivariable Regression Analysis for Liberation Success